[11:15:45] hey y'all [11:15:49] i've been around for a while now [11:15:55] nevertheless, hey y'all [11:15:59] it's good to see you [11:16:29] bmansuro_: my rec yesterday wasn't your thing, but don't give up on me just yet [11:16:35] u137 - dreamer on the run [11:17:06] some good ol' shoegaze/post-rock [11:17:41] https://www.youtube.com/watch?v=zU7beUagj_w [11:17:50] let's see [11:27:54] pretty good [11:28:57] the album is varied but equally as lovely [11:34:37] excellent [17:03:01] joakino, jdlrobson: retro? [17:37:16] how much would you agree with this statement? [17:37:19] ""Bootstrap is the most popular HTML, CSS, and JS framework for developing responsive, mobile first projects on the web."" [17:39:14] 75% [17:45:42] bmansurov: thanks :) [17:46:14] etonkovidova: Could you attend iOS standups if they were M/W/F @ 1030? Calendar seems to agree but wanted to check in. [17:47:37] mbinder: yes, I think so. Recently Collaboration team decided to move their standups to 10:45 - so 10:30 is fine [18:22:11] mutante: context? :) [18:27:47] mutante: it is definitely the most popular, so 100% [18:28:27] jdlrobson: somebody suggests to add it and use it on a labs project i'm in and i wanted to get some opinion on bootstrap itself [18:28:31] joakino: thanks [18:28:53] yeh that sounds pretty common and is exactly why violetto / Volker_E in ux standardisation were exploring ways to support these community members [18:30:04] jdlrobson: ok:) i'll take it then [18:55:46] back from getting the kids to bed [18:56:57] mutante we have wikimedia styled bootstrap here: wikimedia-ui.wmflabs.org [19:11:05] jdlrobson: https://phabricator.wikimedia.org/T113243#1898854 "What can we help with?" ? :) A review is always welcome :P [19:16:55] FlorianSW: already on that [19:17:03] was curious why WIP though [19:17:06] seems quite far along [19:18:08] jdlrobson: because I got no feedback about the todo in the commit message so far :) [19:18:27] let me post my review 1s [19:18:51] :) [19:21:01] hello everyone :) [19:23:24] mutante: Hi! [19:24:21] nice work JoshM! [19:26:21] mutante: As long as you're aware of the (highly possible) overhead of Bootstrap and you know what to do with or against that overhead, it can be useful in a time-constraint project cc: jdlrobson [19:27:49] jdlrobson: FlorianSW could I take up one or two tasks from Gather/MobileFrontend tagged as GCI and requiring mentors, to mentor, given that I do not have a +2 on them? [19:28:18] codezee: If you're a mentor in GCI, sure :) [19:28:37] codezee: what are the links i'll take a look [19:28:48] but I think that's fine yes [19:29:04] I'll be on hand to +2 them codezee if necessary [19:29:40] Volker_E what's an example of your concern about bootstrap about its overhead? [19:31:26] mutante: violetto came up with a custom implementation of Wikimedia UI properties in one of her side-projects at https://github.com/munmay/WikimediaUI (you've probably come there already) and we (UI Standardization) are currently discussing going to make Bootstrap (due to its popularity; also on wmflabs) an example implementation in our further work [19:33:17] bgerstle, hai [19:33:24] dtm hey! [19:33:28] :-o [19:33:39] so did you see the beta announcement email? also, are you in the community or WMF? [19:34:07] violetto: As I've said on several occasions IRL, but not yet written down. It's huge for a front-end library and the minority of developers don't implement it in a modular way. The just shoot the whole library down the wire at the clients with a negative performance impact [19:34:26] violetto: repeating "not yet written down" :) [19:35:25] dtm: sorry forgot to mention, see my message above [19:37:08] Volker_E like a few people have mentioned above, bootstrap is very common. It's very likely someone has visited a site with bootstrap, so it's already cached in a user's browser so any subsequent requests to download the file is instantaneous and requires zero bandwidth [19:39:52] violetto: First of all, the dev has to be aware of cache settings and set them usually right in order to get the optimal speed out [19:41:12] violetto: More important caching on mobile is in various cases still sloppy and you also lose a lot of speed on first load (which might result in losing visitors) [19:41:18] whatever the cache settings are, it's going to be cached, no? [19:41:33] also, the initial download is smaller than a tiny image [19:41:42] I would agree that modularity is ideal if devs are using only one button, for example, in what they're building. That is assuming that this type of use case is frequent. [19:42:00] violetto: There has been long-time issues with mobile browsers and caching in certain cases [19:43:23] jdlrobson: thanks :) and something like https://phabricator.wikimedia.org/T114389 [19:43:54] hence I think we should find out how our users will be using the library before we forgo a potentially good solution [19:46:35] Deskana|Away: you're wrong, I want "Electric Eel (Dungeons & Dragons" [19:47:14] violetto: In contrast to images, CSS is render blocking https://developers.google.com/web/fundamentals/performance/critical-rendering-path/render-blocking-css [19:47:58] bgerstle, what community? i'm not in the WMF so i guess yes! [19:48:40] so it does matter how big a CSS file is and how/how fast it's sent down the wire [19:49:16] dtm that answers my question [19:49:21] bgerstle, in my opinion, one strong aspect and use case of mobile computing is casual browsing. like i said before, that tends to dominate and dumb it all down. but. you're talking about being so casual that you get a "random article" button. we need a personal portal, dude. you need to bust out your phone and check your wikipedia dashboard and portal. [19:49:26] bgerstle, https://en.wikipedia.org/wiki/User:Smuckola#Wikipedia_wishlist [19:49:32] dtm send me an email (bgerstle@wikimedia.org) w/ your AppleID and i'll make sure you're added to the beta list [19:50:12] dtm we're taking first steps there w/ the feed and would _really_, _really_ love your feedback on how it serves that purpose [19:50:22] Deskana|Away: be sure to test with Uneited Kingdom too ;) [19:50:23] jdlrobson: hello. I have no clue how you manage to fix qunit tests. But your last patch on MobileFrontend is awesome :-) [19:50:30] hashar: lol [19:50:36] the patch that fixes the qunit tests? [19:50:42] jdlrobson: yeah. [19:50:44] but ultimately it sounds like what you're after is collections (personal curation) which Android will start exploring next quarter [19:50:44] violetto: I hope you saw my last msg [19:50:49] jdlrobson: the one preventing additional api calls on scrolling [19:50:52] bgerstle, you need to be able to see a quasi-realtime feed of the state of My Wikipedia. it's what's happening that i'm interested in, like my watchlist, my vandalism report, friend list, new users who've started contributing to my watchlist [19:51:07] both for editing and reading [19:51:09] hashar: WAs interested that it only seemed to be triggering a scroll randomly [19:51:26] jdlrobson: I had absolutely no clue what I was actually debugging nor whether the log I was pasting would made any sense. I am quite happy it is a one-liner fix that caused headaches for a while :D [19:51:26] dtm we also have a public backlog board, where you can file & track these feature requests https://phabricator.wikimedia.org/tag/wikipedia-ios-app-product-backlog/board/ [19:51:29] and any bugs ;-) [19:51:35] please drop tickets into needs triage [19:51:46] bgerstle, and i need not just a watchlist but a todo list. i want to pin up (with notes) the articles i'm wanting to work on when i get back home (not mobile) [19:51:50] hashar: yeh qunit blaming VE was not helping [19:51:57] dtm yeah, that'd be sweet [19:52:04] bgerstle, so obviously what i'm talking about is a mobile-nonmobile overlap [19:52:16] jdlrobson: we would need better debug output and console logging to be added to the qunit jobs. That would ease diagnostic on test failure [19:52:19] bgerstle, there's no reason this needs to be mobile but it's a universal feature that appeals strongly to mobile [19:52:23] dtm yeah, we talk about that a lot as well: cross-platform cohesiveness [19:52:25] if you're looking for mobile things [19:52:47] it's a universal thing that helps you be more mobile, and helps you not lose things when you're mobile [19:52:53] but everyone needsit [19:53:13] dtm: preaching to the choir, my friend :-) [19:53:21] bgerstle, basically i'm talking about breaking down wikipedia culture's stone-brained denial of the fact that it is indeed a social network. [19:53:30] make sure you send me that email and i'll get you on the beta. we love hearing from our users [19:53:41] word [19:54:00] dtm not sure if you saw, but someone from the community-engagement/tech team responded to you in #wikimedia-tech [19:54:20] oh. [19:54:25] definitely keep in touch w/ us (on the iOS app team), but definitely make sure to use those channels for higher-level feedback about the org [19:54:30] @Volker_E it's probably worth a real life comparison on bootstrap vs non-bootstrap loading time. It could be a little larger (due to bootstrap's comprehensive library vs others), but my hunch is that it's still instantaneous. We should compare load time and weigh it against the *advantages* of Bootstrap. [19:54:42] Volker_E: "There has been long-time issues with mobile browsers and caching in certain cases" <-- sounds like an app issue rather than bootstrap [19:55:31] dtm great to meet you though! i have a couple meetings coming up now so i'll be less responsive, but hope to be in touch again soon once you have the 5.0 beta :-) [19:55:48] violetto: I think that Bootstrap has it's use cases, but it's far from being a general solution as it comes with issues like I laid out before [19:56:17] i'm not even super sure with that yet :) [19:56:46] but im happy to provide that solution while we think of others / asses bootstrap as a solution [19:57:33] because while we cannot promise to support, bootstrap has a community to help with that support that our users need. [19:57:41] violetto: what are you not sure about yet? [19:57:54] if this is *the* general solution [19:58:25] violetto: don't understand, "this" what? [19:59:42] ? [20:00:09] violetto: what are you not sure about yet? [20:00:31] violetto: write a full sentence without this/that ;) [20:01:36] "but it's far from being a general solution as it comes with issues like I laid out before" <-- I'm not super sure with this statement of yours [20:02:29] dtm: have you tried Gather? [20:02:34] violetto: so none of my arguments above with the links connected made sense to you? [20:04:04] violetto: Most of the work that Performance/Reading mobile team is doing (and the web in general should go to implement) is breaking down, modularizing, minifying, squeezing out whereever. Squeeze out every little bit for better performance as there is clear statistics that it helps increasing recurring visitors, visitor satisfaction and (in eCommerce sites) [20:04:04] conversions [20:04:18] https://en.m.wikipedia.org/wiki/Special:Gather/all/recent < dtm [20:05:01] violetto: As I said, Bootstrap has it's valid use cases, like for example prototyping. [20:05:39] Volker_E: it sounds like May is talking about bootstrap on labs which makes sense, where performance is less of a concern [20:05:40] it makes very much sense, but they seem to be minute details compared to what bootstrap offers [20:06:10] bootstrap is great for getting ideas off the ground which is what labs is about - creativity and making ideas happe [20:06:17] * happen fast [20:06:27] violetto: jdlrobson I never doubted it, as long as the dev knows what is the output [20:06:41] jdlrobson: violetto: and what the possible shortcomings are [20:07:56] Volker_E: i also hope you understood my arguments although you're not 100% with bootstrap. [20:08:50] i'm open with other solutions depending on what our users need, and I hope you can do the same. our own new solutions are always tempting, but there are sometimes other open source solutions we can take advantage of [20:09:39] fyi it's not 100% clear to me what you two are are arguing about. Devs on labs will use tools they are familiar with. For instance I made http://pushipedia.wmflabs.org without mediawiki or oojs ui in plain node with an html and css file. Providing stylesheets that can be slotted into any framework would be highly useful, especially Bootstrap since that seems [20:09:39] to be used a lot. I hope this is somewhat helpful. [20:10:38] violetto: I do. As a general approach (for all web projects out there) I heartfully and from my expertise disagree, but we should look into further inputs from others. No doubt on that as well. ;) [20:10:40] i was responding to mutante and bootstrap [20:10:55] me too [20:11:17] bootstrap on labs to be specific [20:11:30] Volker_E: yup [20:11:39] im glad we at least agree on that [20:11:44] wanted to let him know (after he said, that he will use Bootstrap) that there is work done by violetto which could greatly fit his needs [20:12:22] violetto: jdlrobson: then kinda drifted into a how much of a solution is Bootstrap in general [20:12:52] violetto: jdlrobson: which is good, I might use my collected arguments somewhen later pro and contra :P [20:31:13] jdlrobson: https://gerrit.wikimedia.org/r/#/c/206490/ Wooohooo! :)) [20:37:44] violetto: Here's one article on the caching issues on mobile (dates back to : https://www.mobify.com/blog/smartphone-localstorage-outperforms-browser-cache/ [20:47:56] jdlrobson, hi. i was editing my wishlist. no i have never heard of Gather. [20:48:06] bgerstle, it's gr8 to meet you too, thanks [20:59:05] Volker_E: it took a while :) [21:25:55] thank you Volker_E and violetto [21:29:14] Volker_E: do you think that caching issue is bootstrap-specific, e.g, would it not happen with other libraries? [21:29:58] mutante: we're happy to hear your feedback... :) [21:31:30] violetto: That's the reason that people are asking for a more modular way in OOjs [21:32:55] jdlrobson, Gather? what in the world is that, lol? it looks funky [21:33:02] violetto: you're right, it's a common problem with libraries getting bigger and fulfilling (too) many things at once.